Skip to contents

Describe how to (re)construct a subject's dataset

Source: R/fmri_job.R
dataset_spec.Rd

A small, serializable recipe for a dataset: the name of a dataset constructor plus the arguments to call it with. For file-backed data the arguments are paths and run lengths (no voxel data), which keeps the enclosing fmri_job tiny and portable. The dataset is realized lazily on the worker.

Usage

dataset_spec(constructor, args = list(), source = c("file", "inline"))

Arguments

constructor

Name of a frame constructor (a string): one of "nifti_frame", "matrix_frame", "neurovec_frame", or "latent_frame". Resolved at run time, so the data is not loaded when the spec is built.

args

A named list of arguments passed to constructor (for "nifti_frame": scans, TR, run_length, event_table, mask, base_path, ...).

source

Either "file" (paths; nothing loaded until run) or "inline" (data already in args, e.g. a matrix for matrix_frame).

Value

An object of class dataset_spec.

See also

fmri_job(), nifti_frame(), matrix_frame()

Examples

dataset_spec("nifti_frame",
             args = list(scans = c("run-1_bold.nii.gz", "run-2_bold.nii.gz"),
                         TR = 2, run_length = c(200, 200)),
             source = "file")
#> <dataset_spec>
#>   constructor: nifti_frame()  source: file
#>   args: scans, TR, run_length